 |
|
Regular readers of ERCB may recall my comments when reviewing Graphics Programming with JFC (see http://www.ercb.com/brief/brief.0133.html):
It's an editorial blemish not to have provided a minimal bibliography to steer novice readers from the paths of pop techdom towards the high road of computer science..."
Real-Time Rendering, by Tomas Moeller and Eric Haines, remedies that omission. If you're ready for the mathematics of 3D computer graphics, you're ready for Real-Time Rendering. The authors state their objective succinctly:
This book is about algorithms which create synthetic images fast enough that the viewer can interact with a virtual environment.
Real-Time Rendering is a real textbook. The text is medium-to-high scientific English. To read the book, you must possess some familiarity with matrix math, geometry, and trigonometry. Equations abound, as do citations.
The presentation is both professional and scholarly. Moeller and Haines weigh in with a great deal of practical experience and an equal proportion of theoretical familiarity. The treatment is never abstract. Specific graphics subsystems are named and differences between them noted in the discussion. Rendering, not mathematics, is the focus.
The headings of Real-Time Rendering's content-rich chapters are:
1. Introduction
2. The Graphics Rendering Pipeline
3. Transforms
4. Visual Appearance
5. Texturing
6. Special Effects
7. Speed-up Techniques
8. Pipeline Optimization
9. Polygonal Techniques
10. Intersection Test Methods
11. Collision Detection
12. Graphics Hardware
13. The Future
Appendix A. Some Linear Algebra
Appendix B. Trigonometry
Real-Time Rendering is about process, not production. Hardware involved in the graphics pipeline is presented at the conceptual level in Chapter 12, "Graphics Hardware" and most engagingly in 12.2.2 "Case Study: Neon." The metal itself is not mined. If it's blit rates and bus signals you crave, you want a book like AGP System Architecture (also reviewed on ERCB; see http://www.ercb.com/brief/brief.0138.html).
Overall, the editing, design, layout, and printing of this hardbound volume are excellent. Care and attention have been lavished on equation presentation. In addition to copious meticulous geometric diagrams, there are a number of plates relevant to the discussion. One group of plates is color; the rest are gray scale. Several of the gray scales are, in my review copy, unfortunately printed a bit too dark for comfort, but this is a minor flaw compensated for chez nous by a nifty fluorescent ring work light with central magnifying lens.
I learned the basics of rendering by taking off my shoes and socks and counting toes, while working through early OS/2 Presentation Manager development kit documentation. It is gratifying that this sort of approach is not necessary for the modern computer-science student who has access to books like Real-Time Rendering and to some of the 392 reference works cited therein, bless the bibliography.
A few sections of the book are available for download from the web site maintained by the authors; see http://www.acm.org/tog/resources/RTR/.
-- Jack Woehr
| Readability |
|
| Originality |
|
| Organization |
|
| Accuracy |
|
| Consistency |
|
| Depth |
|
| Timeliness |
|
| Editing |
|
| Design |
|
| Overall Value |
|
Explanation of ERCB rating scale: No stars = unacceptable, 1 Star = marginal, 2 Stars = average, 3 Stars = above average, 4 Stars = exceptional.